Project Abstract Understanding how the immune landscape is established and how it influences therapeutic response in breast cancer are critical issues. We hypothesize that epithelial cancer cell-derived, differentially expressed genes contribute to the establishment of distinct tumor-immune environments and impact patient survival and response to immunotherapy. By comparing immune-related expression signatures of ~1800 breast tumors, we identified three distinct immune groups: immune-low, normal, and immune-high. Immune-high patients had significantly longer recurrence-free survival. For the first aim of this proposal, we will validate the clinical relevance and key immunological features associated with these immune groups in additional datasets. Flow cytometry will be used to detect cell type-specific markers for immune cell types associated with these immune classes. Interestingly, comparisons of somatic mutation and neoantigen load show no significant differences between the immune-low and -high groups; whereas, large differences in gene expression exist between these two groups. The influx of immune cells presumably present in immune-high tumors confounds determination of the source of expression from bulk RNAseq data. Therefore, we used expression data from in vitro cell lines and patient-derived xenografts grown in immune-compromised mice as surrogates for epithelial-specific gene expression. This approach was used as a filter to select genes expressed by epithelial tumor cells. The C-C Motif Chemokine Ligand 5 (CCL5), a known chemoattractant for T lymphocytes, was expressed by epithelial cells and had the second highest fold-change between immune-high and -low patients. Thus, CCL5 may be a critical determinant of immune involvement. A more direct interrogation into the source of differential expression is proposed as the second aim of this project, in which single-cell RNAseq of human and mouse tumors is used to identify epithelial cancer cells derived genes that distinguish immune-low and ?high tumors. Single-cell RNAseq defined epithelial- expressed genes will be ranked by immune-high versus -low fold change. The top ranked candidate and CCL5 will be used in a proof-of-principle study in aim three to determine whether modulation of these candidate immune-determining genes is sufficient to increase immune cell presence and improve response to checkpoint inhibition therapy in TP53-/- mouse mammary tumor lines grown orthotopically in syngeneic, immune-competent BALB/c mice. If successful, the results of this project will lead toward the long-term goal of this research, which is to develop a therapeutic strategy that effectively engages the anti-tumor effects of the immune system in patients that have low immune involvement. The cross-disciplinary training and mentorship in cancer biology, bioinformatics, mouse models, and immunology that have been combined in this proposal highlight the impactful work being done at UNC and will provide immense potential for developing a successful independent researcher.